The Mechanics of Standard-Essential Patents: How FRAND Licensing and Royalty Stacking Actually Work

The fundamental tension at the heart of modern technology is a paradox of cooperation and competition. For a smartphone to connect to a cellular network anywhere in the world, competing manufacturers must agree on a single, universal standard. Yet, the underlying technologies that make up that standard—from signal encoding to error correction—are proprietary, owned by the very companies competing for market share. Resolving this tension requires a legal fiction that has become the bedrock of the global tech economy: the Standard-Essential Patent (SEP) and its accompanying FRAND commitment.[4]

When a standard-setting organization (SSO) like the IEEE or 3GPP adopts a new technology, any patent essential to implementing that standard is designated an SEP. Because a manufacturer cannot build a compliant device without infringing these patents, the patent holder gains immense market power. To prevent extortion, SSOs require SEP holders to license their technology on "Fair, Reasonable, and Non-Discriminatory" (FRAND) terms. However, FRAND is a principle, not a formula. It dictates that a license must be offered, but it rarely specifies the exact price, leaving the actual mechanics of valuation to a deeply opaque market.[2]

The evidence surrounding FRAND licensing reveals a stark divide between what is claimed and what is actually paid. Patent holders frequently market their portfolios by citing "bottom-up" royalty rates—assigning a specific percentage of a device's selling price to their specific patents. If every SEP holder for a standard like 5G successfully extracted their stated bottom-up rate, the cumulative royalties would quickly exceed the profit margin of the device itself. This phenomenon, known in the literature as "royalty stacking," is the primary systemic risk in standard-setting.[1][3]

To combat royalty stacking, courts and regulators increasingly rely on "top-down" methodologies. This approach first establishes an aggregate royalty burden for the entire standard—often capped between 5% and 10% of the end product's price—and then allocates that pie among patent holders based on the relative strength and size of their portfolios. The evidence shows that top-down determinations consistently yield significantly lower individual royalty rates than bottom-up demands, forcing a rationalization of the market that marketing materials rarely acknowledge.[3]

Despite the vast sums involved, the empirical evidence on FRAND licensing is remarkably thin. Because the vast majority of SEP licenses are negotiated bilaterally and protected by strict non-disclosure agreements, researchers and policymakers are forced to extrapolate from the small fraction of disputes that end up in public litigation. This creates a selection bias: the data we have represents the most contentious, high-stakes negotiations, rather than the routine, functioning baseline of the industry.[1]

In recent years, the mechanics of SEPs have shifted from corporate boardrooms to national policy agendas. As control over critical technologies like 5G and artificial intelligence becomes a matter of national security, governments are increasingly treating SEP portfolios as strategic assets. The evidence suggests that jurisdictions are beginning to use their domestic courts to set global FRAND rates, attempting to advantage their own domestic champions by either inflating the value of their patents or depressing the royalties they must pay to foreign competitors.[4]

The academic literature is sharply divided on the primary market failure in SEP licensing. One camp points to "patent hold-up," where SEP holders use the threat of an injunction to extract exorbitant royalties after a manufacturer is locked into a standard. The opposing camp emphasizes "patent hold-out," where implementers deliberately delay negotiations and force patent holders into expensive, protracted litigation, knowing the ultimate court-determined rate will likely be lower than the initial demand. The evidence supports the existence of both behaviors, though their relative prevalence remains fiercely debated.[1][2]

Ultimately, the FRAND system, despite its inefficiencies and the constant threat of litigation, functions well enough to enable the modern interoperable world. The sheer volume of connected devices shipping daily is evidence that, while the exact mechanics of royalty calculation remain contested, the baseline incentive structure holds. Companies continue to invest billions in research and development to contribute to standards, and manufacturers continue to license those standards to build products. The friction is a feature of the system's design, not a bug that threatens its collapse.[5]